[time-nuts] 10MHz LTE-Lite

Bob Camp kb8tq at n1k.org
Mon Dec 8 19:42:30 EST 2014


On the 5335, the display changes a bit as you fiddle with the math. The actual resolution does not change. When you get an extra digit, it really does not step off in one count steps. 

Anything you do that gets you to a 10X longer gate time does add a “real” digit. Taking 10 back to back readings increases the display by about 1/3 of a digit (back to the square root N issue). 


> On Dec 8, 2014, at 6:09 PM, Keith Loiselle <keith.loiselle at gmail.com> wrote:
> Enabling MEAN with 100 pts under Statistics should give an additional digit
> but will take a few minutes for a reading with longer gate times.
> Keith
> Keith
> On Mon, Dec 8, 2014 at 3:07 PM, Bob Camp <kb8tq at n1k.org> wrote:
>> HI
>> At a 1 second gate, your 5335 is good to about 1 ppb. (1x10^-9). A TCXO
>> based GPSDO should be good to 10X that level. An OCXO based unit should be
>> good to 100X that level.  If you extend the counter’s gate time, the 5335
>> will overflow fairly quickly. Up to the point it does, it’s accuracy will
>> improve directly with the gate time. Once it overflows, you can correct the
>> result, but it is messy.
>> Bob
>>> On Dec 8, 2014, at 2:07 PM, Byron Hayes Jr <bhayes at earthlink.net> wrote:
>>> Time-Nuts Group,
>>> I thought some of you might be interested in my experience with the
>> 10MHz LTE-Lite.
>>> The 10MHz LTE-Lite arrived about a week ago.  I was not ready to make a
>> permanent installation, I wanted it portable and I wanted to get started
>> quickly.  I am hobbyist interested mainly in the HF spectrum.  So I decided
>> to operate the LTE-Lite inside the Priority Mail box, which was pretty much
>> intact.  I cut a 5" X 8" piece of corrugated cardboard, mounted the unit on
>> it with two doublestick pads, and put the cardboard with the unit into the
>> Priority Mail box.  I cut a hole in the top of the box so I could see the
>> LEDs, and cut three small holes in the side, one for the USB cable, one for
>> the antenna wire and one for the 10 MHz output.  I attached the USB cable,
>> antenna wire and 10 MHz output cable to the unit and ran them out of the
>> box through the holes.  I had an operating Lenovo X220 Windows 7 computer
>> near to the box, so I plugged the USB cable into the computer (I did not
>> try to get or use any software on the computer to decipher any messages
>> from the LTE-Lite).  I was in a
>> n upstairs North facing bedroom (in the Los Angeles area) so I put the
>> antenna on a nearby windowsill.  I hooked the 10MHz output to the channel 1
>> (Hi Z) input of my Tektronix 222A osciloscope, with the trigger on channel
>> 1.
>>> When power was applied through the USB line, the LEDs seemed to light
>> normally.  Within a couple of hours the lock LED was on, but the
>> oscilloscope was showing noise, not a meaningful output.  I let the whole
>> thing sit overnight, and the next day an apparent 10 MHz trace was on the
>> screen.  It was not a sine wave, and not a square wave, but something in
>> between.
>>> I had a small Rb unit, an Efratom 10 MHz FRS-C built into a TM-500
>> plug-in.  I set that up and let it warm up and lock.  I connected the Rb
>> output to channel 2 of the 222A and got that trace on the screen.  It was a
>> sine wave basically in lock step with the LTE-Lite trace.  Over a few hours
>> one could see slight relative movement, but very slight.  What next?
>>> I had a couple of HP 5300 series frequency counters, one a 5300B display
>> with Option 1 (Hi-Stability time base) and a 5308B lower unit, and the
>> other a plain 5300B display with a 5303B Option 1 (Hi-Stability time base)
>> lower unit.  I give both time to warm up.  I put a "T" in the LTE-Lite
>> output line and another "T" in the Rb output line, and connected coax from
>> the Ts to the HiZ input of each frequency counter.  After they settled
>> down, the counter connected to the LTE-Lite read 10000008 and the counter
>> connected to the Rb read 10000002.  So, after a while, I reversed the leads
>> to the counters, and the counter connected to the LTE-Lite read 10000002
>> and the one conntected to the Rb read 10000008.  Those readings have been
>> consistent for several days.  That indicated to me that the LTE-Lite and
>> the Rb were both outputing essentially the same frequency, but the counters
>> were a bit off (I had never calibrated these counters, since I wasn't sure
>> of the accuracy of the Rb unit).  But,
>> I felt like the proverbial man with two watches.
>>> So, I brought out "Big Gonzo", a HP 5335A counter with Option 010
>> (Hi-Stability time base) I purchased on eBay about six months ago but had
>> never fired up.  It came up OK and I hooked it to the LTE-Lite output.  It
>> initially read 10000028, but I gave it a couple of hours to warm up the
>> oscillator oven and stabilize.  By then, the reading had settled to 10 000
>> 000.  Now I switched cables and hooked the Rb to the 5335A, and it read 10
>> 000 000.  I hooked the LTE-Lite to channel A of the 5335A and the Rb to
>> channel B of the 5335A, and set the counter to "ratio".  When it settled,
>> the counter read 1.000 000 indicating that the two outputs were the same
>> frequency, at least to six decimals.  Apparently the 5335A is "right on".
>>> I conclude that both the LTE-Lite and the Rb are outputting a 10000000
>> MHz signal.  I'm not sure of the accuracy beyond that.  They are probably
>> both accurate enough for my purposes for calibrating other test equipment,
>> receivers and transmitters operating in the HF spectrum.
>>> I would be interested in any comments or suggestions from other list
>> members.
>>> Byron WA6ATN  _______________________________________________
>>> time-nuts mailing list -- time-nuts at febo.com
>>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>>> and follow the instructions there.
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.

More information about the time-nuts mailing list